DP01: Getting the whole picture on melanoma diagnosis: investigating the effects of clinical images and dermoscopy report on the histopathology diagnosis of melanocytic skin lesions

Orals

Belinda Lai,¹ Blake O’Brien,² Tristan Dodds,^3,4 Peter Ferguson,^1,5,6 Richard Scolyer,^1,5,6 Gerardo Ferrara,⁷ Giuseppe Argenziano,⁸ H. Peter Soyer^9,10 and Katy Bell¹

¹The University of Sydney, Sydney, Australia; ²Sullivan Nicholaides Pathology, Brisbane, Australia; ³Laverty Pathology, Sydney, Australia; ⁴The University of California San Francisco, San Francisco, CA, USA; ⁵The Royal Prince Alfred Hospital, Sydney, Australia; ⁶Melanoma Institute Australia, Sydney, Australia; ⁷Macerata General Hospital, Macerata, Italy; ⁸The University of Campania Luigi Vanvitelli, Caserta, Australia; ⁹The University of Queensland, Brisbane, Australia; and ¹⁰Princess Alexandra Hospital, Brisbane, Australia

While setting the reference standard diagnosis for 98 melanocytic lesions excised from patients attending a dermatology clinic in Italy from January 2004 to December 2005, we investigated differences in four expert dermatopathologists’ interpretations of the histopathology. We aimed to assess the effects of providing clinical images of melanocytic lesions (macroscopic and dermoscopic) and, additionally, a dermatologist’s report on the histopathology diagnoses made for each case. We report results for the four pathologists’ interpretations of the first 40 melanocytic lesions. The expert dermatopathologists each provided an independent review of the cases through an online platform. For each case, the pathologist provided interpretations for three sequential reading conditions: (i) histopathology only; (ii) histopathology, macroscopic and dermoscopic image; and (iii) histopathology, macroscopic and dermoscopic image, and dermoscopy report. At each reading condition, the pathologist assigned a histopathological diagnosis using the Melanocytic Pathology Assessment Tool and Hierarchy for Diagnosis (MPATH-Dx) classification scheme and indicated their level of diagnostic confidence on a 5-point scale ranging from 1 (no diagnosis can be made) to 5 (no other diagnosis is considered). The 40 melanocytic lesions were from 39 patients (19 males, 20 females, mean age 40·9 years). The original histopathological diagnosis provided to the patient was naevus for 16 lesions and melanoma for 24 lesions. There were 160 interpretations for the 40 lesions (four pathologists × 40 lesions) for each reading condition. Twenty interpretations of the histopathology alone were MPATH-Dx class I; five were then reclassified as MPATH-Dx II and one as MPATH-Dx III with clinical images and dermoscopy report (14 remained MPATH-Dx I). Fifty-four interpretations of histopathology alone were MPATH-Dx class II; 14 was reclassified as MPATH-Dx III (40 remained MPATH-Dx II). Thirty-nine interpretations of histopathology alone were MPATH-Dx class III; two were reclassified as MPATH-Dx V (37 remained MPATH-Dx III). Twenty-five interpretations of histopathology were MPATH-Dx IV and 27 were MPATH-Dx V; none was reclassified with clinical images and dermoscopy report. The mean level of diagnostic certainty increased a modest amount when clinical images and dermoscopy report were provided (from 3·6 for histopathology alone to 3·8 with clinical images and dermoscopy report). The pathologists indicated they would like a second opinion on 20 of the 160 interpretations based on histopathology alone: two MPATH-Dx I, 11 MPATH-Dx II, four MPATH-Dx III, one MPATH-Dx IV and two MPATH-Dx V. After the provision of clinical images and dermoscopy report, this increased to 25 interpretations: one MPATH-Dx I, 11 MPATH-Dx II, nine MPATH-Dx III, one MPATH-Dx IV and three MPATH-Dx V. In conclusion, providing clinical images and a formal dermoscopy report may influence the diagnostic certainty and histopathology diagnosis, particularly for lesions assigned to MPATH-Dx class II based on histopathology alone.